When it comes to wavefront sensing of the human eye, there does not seem to be an agreed set of specifications for detector selection. In my career, camera choice seem to rely on legacy more so than anything else. The earliest labs bought really expensive cameras (i.e. COHU) which eventually became obsolete.
New researchers joining the lab later on either use the old cameras if they still worked or buy the new model that the same manufacturer intended to replace the older model. Since the obsolete model was typically top-of-the-line, a suggested new model is also often the case (along with the hefty price tag).
Perhaps we can all save some money by establishing specific criteria and pick the cheapest camera that does the job? What are your requirements and why?
Well at my university, some H-S sensors have been built with 25US$ web-cams...
further, also there some of those H-S sensors have a "lenslet" that has been just printed out by the students...
Both for measuring aberrations of the human eye
So very, very cheap is possible...
Now the question is what do you want to address and measure, accuracy, resolution, SNR, minimum light levels, etc... are some of the elements that may drive decision towards more expensive systems
Exactly Samuel. That's the type of discussion I'm trying to get at. I would like to attach some numbers to those parameters. I'm actually only interested in ocular wavefront sensing A $25 web cam will most likely not work because it might not be all that sensitive, and that's important because there's only so much light you can introduce into the human eye safely, and for some eye you barely get any of it back. What do you think?
I can tell you the 25US$ Web-Cam works for measuring the aberrations of the living human eye, but you can only have a few elements in your lenslet array (say below 49 elements, 7x7 array).
You get more photons because every spot illuminates a higher number of pixels, further you can work also with the integration time.
Again, accuracy, resolution, SNR are compromised.
I agree with Sam in some of his comments. The question is the SNR of the spots. You need small spots with lots of photons, thus microlenses with high numerique aperture. You need high SNR to get spots above the background level and then segment them , label it and compute their centroids (after removing the background). Also important is the wavelength, for infrarred you can use a cheap CMOS of 8 bits. There is a very sensitive webcam often used by astronomers the Philips Toucam (i think thats the correct spelling) I think it is sensitive enought and is near 100 euros.
Unfortunately, I don't have anything scientific to say on the subject, but I still recognize this is an important piece of wavefront sensing work that is still missing in the community. I'm currently working on a document explaining wavefront reconstruction for ocular wavefront sensing from 1st principles. That is - everything calculation step going from spot locations in Shack-Hartmann images to some representation of the wavefront error (including by not limited to Zernike coefficients). I think it would be a good paper for the upcoming Applied Optics' Engineering and Laboratory Notes. By my observation, there's too many papers that would say that they measured aberrations with a wavefront sensor and boom! here are the Zernike coefficients. When people tell me that, I often interrupt them and ask if they used the algorithm described in Cubalchini's seminal paper in 1979. So far, no one even knew who Cubalchini was. We should all remember that if input data is flawed, no amount of analysis will produce a meaningful result.
J Arines , you mentioned some CMOS sensor suitable for IR wavelength, could you suggest the model. I am involve SHS IN IR band 3- 5 um.
Hellp Sanjay,
At that range I don't know any CMOS sensor. But I think that CMOS technology would not be able to operate ata that wavelengths. See the next link as an example
Probably at that range you would need a more expensive solution.
Justo
http://www.google.es/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&cad=rja&uact=8&ved=0CDMQFjAE&url=http%3A%2F%2Fwww.aphesa.com%2Fdownloads%2Fdownload2.php%3Fid%3D1&ei=0auCVIPLKcj4Uq_cg-gP&usg=AFQjCNG9S8Vix59mZKa5fYeZ-mvSw0ZaOQ&sig2=6MGAbBzcFODCR_teyfI0-A&bvm=bv.80642063,d.d24
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Medical Pharmacology
- Clinical Pharmacology